Pub Date : 2020-06-05DOI: 10.1109/TENSYMP50017.2020.9230731
Tanbin Islam Rohan, Md. Salah Uddin Yusuf, Monira Islam, Shidhartho Roy
Epileptic seizure is one of the common neurological disorder now a day. But this is curable if it can be detected in the early stage. So, this research become a necessity in the early prediction of epileptic seizure. A complete and reliable system can classify the epileptic seizure patients and the states of epileptic seizure. This research explores a supervised machine learning and deep learning model for the classification of epileptic seizure patients from the Epileptic Seizure dataset of UCI machine learning repository. The dataset has 11,500 instances; every information contains 178 attributes. XGBoost is used for the Machine learning approach and ANN is used for Deep learning approach. The proposed ANN algorithm has the improved accuracy and accurately classified the epileptic seizure class patients. 10-fold cross validation is used for the validation purpose. XGBoost acquires 96.6% test accuracy and ANN acquires 98.26% test accuracy. The proposed Deep Learning approach has out-performed the conventional epileptic seizure classifier algorithms. Additionally, the Deep learning model enhances the performance of epileptic seizure detection.
{"title":"Efficient Approach to Detect Epileptic Seizure using Machine Learning Models for Modern Healthcare System","authors":"Tanbin Islam Rohan, Md. Salah Uddin Yusuf, Monira Islam, Shidhartho Roy","doi":"10.1109/TENSYMP50017.2020.9230731","DOIUrl":"https://doi.org/10.1109/TENSYMP50017.2020.9230731","url":null,"abstract":"Epileptic seizure is one of the common neurological disorder now a day. But this is curable if it can be detected in the early stage. So, this research become a necessity in the early prediction of epileptic seizure. A complete and reliable system can classify the epileptic seizure patients and the states of epileptic seizure. This research explores a supervised machine learning and deep learning model for the classification of epileptic seizure patients from the Epileptic Seizure dataset of UCI machine learning repository. The dataset has 11,500 instances; every information contains 178 attributes. XGBoost is used for the Machine learning approach and ANN is used for Deep learning approach. The proposed ANN algorithm has the improved accuracy and accurately classified the epileptic seizure class patients. 10-fold cross validation is used for the validation purpose. XGBoost acquires 96.6% test accuracy and ANN acquires 98.26% test accuracy. The proposed Deep Learning approach has out-performed the conventional epileptic seizure classifier algorithms. Additionally, the Deep learning model enhances the performance of epileptic seizure detection.","PeriodicalId":6721,"journal":{"name":"2020 IEEE Region 10 Symposium (TENSYMP)","volume":"19 1","pages":"1783-1786"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77723907","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/TENSYMP50017.2020.9230640
Shamudra Dey, Tahsin Ibtida, C. Roy, Nuruzzaman Sakib
A computational analysis is performed using the Large Eddy Simulation (LES) method to predict non-Newtonian pulsatile flow of blood into Artery Bypass Graft (ABG) models with varying anastomosis angles. The traditional end-to-side bypass graft models of three distinct anastomosis angles (30°, 45° & 60°) are created on the 75% idealized arterial stenosis to serve the purpose of vascular reconstruction. A pulsatile parabolic inlet boundary condition with the contemplation of spiral component is formulated to model the naturalistic flow in the blood vessel & non-Newtonian Carreau viscosity model is used. The transient study was performed in the ANSYS Fluent 15.0. Initially, the numerical simulation process has been validated with the results of a classic experimental study and Direct Numerical Simulation (DNS) study. The numerical study found that wall shear stress has significant effect on re-blocking while pressure has smaller effect. The artery bypass configuration of 30° anastomosis angle is the most favorable one for the common end-to-side graft design. As a whole, the LES study has provided basic but novel observations of flow projection in the specific artery bypass graft (ABG) models.
{"title":"Effect of Varying Anastomosis Angles for Non-Newtonian Pulsatile Blood Flow through Artery Bypass Graft Models: an LES Study","authors":"Shamudra Dey, Tahsin Ibtida, C. Roy, Nuruzzaman Sakib","doi":"10.1109/TENSYMP50017.2020.9230640","DOIUrl":"https://doi.org/10.1109/TENSYMP50017.2020.9230640","url":null,"abstract":"A computational analysis is performed using the Large Eddy Simulation (LES) method to predict non-Newtonian pulsatile flow of blood into Artery Bypass Graft (ABG) models with varying anastomosis angles. The traditional end-to-side bypass graft models of three distinct anastomosis angles (30°, 45° & 60°) are created on the 75% idealized arterial stenosis to serve the purpose of vascular reconstruction. A pulsatile parabolic inlet boundary condition with the contemplation of spiral component is formulated to model the naturalistic flow in the blood vessel & non-Newtonian Carreau viscosity model is used. The transient study was performed in the ANSYS Fluent 15.0. Initially, the numerical simulation process has been validated with the results of a classic experimental study and Direct Numerical Simulation (DNS) study. The numerical study found that wall shear stress has significant effect on re-blocking while pressure has smaller effect. The artery bypass configuration of 30° anastomosis angle is the most favorable one for the common end-to-side graft design. As a whole, the LES study has provided basic but novel observations of flow projection in the specific artery bypass graft (ABG) models.","PeriodicalId":6721,"journal":{"name":"2020 IEEE Region 10 Symposium (TENSYMP)","volume":"2 1","pages":"1494-1497"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82912258","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/TENSYMP50017.2020.9230657
Sriparna Banerjee, Shambhab Chaki, Soham Jana, S. S. Chaudhuri
This paper introduces a novel Refined Color Channel Transfer (RCCT) prior as an improved alternative of existing Color Channel Transfer (CCT) prior. Like CCT, RCCT also compensates the chromatic losses occurring in degraded hazy images by employing a global color transfer strategy but it performs color transfer using well-scaled reference images generated using our proposed Fuzzy logic based reference image generation technique in contrary to CCT which usually performs color transfer using reference images possessing over-enhanced glow (bright) regions and poorly enhanced lowlight regions. The presence of such over-enhanced /poorly enhanced regions in the references images used by CCT significantly affect the visibility of outputs obtained from the dehazing methods where CCT acts as a pre-processing step. To overcome these shortcomings, here we have proposed a novel Fuzzy logic based reference image generation technique which restricts the intensities of generated reference images within allowable ranges by introducing a control parameter ‘k’. A unique value of ‘k’ used for controlling the intensity of each pixel is computed depending upon the properties of the super-pixel in which it belongs, using a novel set of Fuzzy Inference (FI) rules which facilitates the production of visually improved outputs and also enables RCCT to serve as an ideal preprocessing step of various daytime, nighttime and underwater dehazing methods which is experimentally proven in this work.
{"title":"Fuzzy Logic-Refined Color Channel Transfer Synergism based Image Dehazing","authors":"Sriparna Banerjee, Shambhab Chaki, Soham Jana, S. S. Chaudhuri","doi":"10.1109/TENSYMP50017.2020.9230657","DOIUrl":"https://doi.org/10.1109/TENSYMP50017.2020.9230657","url":null,"abstract":"This paper introduces a novel Refined Color Channel Transfer (RCCT) prior as an improved alternative of existing Color Channel Transfer (CCT) prior. Like CCT, RCCT also compensates the chromatic losses occurring in degraded hazy images by employing a global color transfer strategy but it performs color transfer using well-scaled reference images generated using our proposed Fuzzy logic based reference image generation technique in contrary to CCT which usually performs color transfer using reference images possessing over-enhanced glow (bright) regions and poorly enhanced lowlight regions. The presence of such over-enhanced /poorly enhanced regions in the references images used by CCT significantly affect the visibility of outputs obtained from the dehazing methods where CCT acts as a pre-processing step. To overcome these shortcomings, here we have proposed a novel Fuzzy logic based reference image generation technique which restricts the intensities of generated reference images within allowable ranges by introducing a control parameter ‘k’. A unique value of ‘k’ used for controlling the intensity of each pixel is computed depending upon the properties of the super-pixel in which it belongs, using a novel set of Fuzzy Inference (FI) rules which facilitates the production of visually improved outputs and also enables RCCT to serve as an ideal preprocessing step of various daytime, nighttime and underwater dehazing methods which is experimentally proven in this work.","PeriodicalId":6721,"journal":{"name":"2020 IEEE Region 10 Symposium (TENSYMP)","volume":"275 1","pages":"654-657"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90781241","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/TENSYMP50017.2020.9230874
S. Arman, Md. Mahfujur Rahman, Syeda Fabliha Rahman, Nazia Parvin Urmi, Progya Paromita Urmee, Nasif Muslim, S. Islam
As the technologies are improving every day, trillions of Internet of Things (IoT) devices are estimated to be deployed in the next five years. The challenge of connecting IoT devices are massive network traffic, high reliability, and energy constraints. In this paper, a Software-defined Network (SDN) testbed with Raspberry Pi which works as an OpenFlow switch is developed. We measure the network parameters (e.g. bandwidth, jitter, throughput) on the testbed without a load balancer and with an active load balancer. The results confirm that SDN based load balancer is useful to overcome those aforementioned challenges. This testbed will be helpful for further research on load balancing in the software-defined network.
{"title":"Developing an IoT Networks-based Testbed for Software-Defined Networks","authors":"S. Arman, Md. Mahfujur Rahman, Syeda Fabliha Rahman, Nazia Parvin Urmi, Progya Paromita Urmee, Nasif Muslim, S. Islam","doi":"10.1109/TENSYMP50017.2020.9230874","DOIUrl":"https://doi.org/10.1109/TENSYMP50017.2020.9230874","url":null,"abstract":"As the technologies are improving every day, trillions of Internet of Things (IoT) devices are estimated to be deployed in the next five years. The challenge of connecting IoT devices are massive network traffic, high reliability, and energy constraints. In this paper, a Software-defined Network (SDN) testbed with Raspberry Pi which works as an OpenFlow switch is developed. We measure the network parameters (e.g. bandwidth, jitter, throughput) on the testbed without a load balancer and with an active load balancer. The results confirm that SDN based load balancer is useful to overcome those aforementioned challenges. This testbed will be helpful for further research on load balancing in the software-defined network.","PeriodicalId":6721,"journal":{"name":"2020 IEEE Region 10 Symposium (TENSYMP)","volume":"49 1","pages":"1752-1755"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90853854","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/TENSYMP50017.2020.9230963
Md. Abu Bakar Siddik, Md. Selim Hossain, A. Paul, Md. M. Rahman, Md. Hassanul Karim Roni, Kisalaya Chakrabatri
In this paper, we demonstrate and evaluate the performance of air filled spiral design photonic crystal fiber (PCF) temperature sensor and double core PCF based temperature sensor. In the first design, selective air holes are filled by temperature sensitive liquid and coated with gold layer to improve the performance of the sensor. In the second design, the high temperature coefficient liquid and plasmonic material are deposited outside portion of the double core PCF to make the fabrication easier. Besides, the coupling phenomenon is studied. The Matlab environment as well as the finite element method (FEM) are utilized to demonstrate the sensor performance. Variation of temperature leads different loss spectra that has been analyzed. The computer simulation result indicate that the obtained wavelength sensitivity of the air filled spiral PCF is as high as 585 pm/°C for y-polarization light and for dual core PCF the sensitivity is increased up to 970 pm/°C and 1075 pm/°C for x and y-polarized light, respectively for wide detection range of temperature 0°C to 80°C. In addition, the variation of structural parameter such as metal thickness and air holes are investigated on the performance of the sensor sensitivity. Considering high sensitivity and low fabrication complexity, the dual core PCF temperature sensor may be a better option to monitor or check the temperature of manufacturing industry, medical environment, transformer oil, battery of electric vehicles and so on.
{"title":"Finite Element Method Based Design Analysis of Internal Coated and External Coated PCF Temperature Sensor","authors":"Md. Abu Bakar Siddik, Md. Selim Hossain, A. Paul, Md. M. Rahman, Md. Hassanul Karim Roni, Kisalaya Chakrabatri","doi":"10.1109/TENSYMP50017.2020.9230963","DOIUrl":"https://doi.org/10.1109/TENSYMP50017.2020.9230963","url":null,"abstract":"In this paper, we demonstrate and evaluate the performance of air filled spiral design photonic crystal fiber (PCF) temperature sensor and double core PCF based temperature sensor. In the first design, selective air holes are filled by temperature sensitive liquid and coated with gold layer to improve the performance of the sensor. In the second design, the high temperature coefficient liquid and plasmonic material are deposited outside portion of the double core PCF to make the fabrication easier. Besides, the coupling phenomenon is studied. The Matlab environment as well as the finite element method (FEM) are utilized to demonstrate the sensor performance. Variation of temperature leads different loss spectra that has been analyzed. The computer simulation result indicate that the obtained wavelength sensitivity of the air filled spiral PCF is as high as 585 pm/°C for y-polarization light and for dual core PCF the sensitivity is increased up to 970 pm/°C and 1075 pm/°C for x and y-polarized light, respectively for wide detection range of temperature 0°C to 80°C. In addition, the variation of structural parameter such as metal thickness and air holes are investigated on the performance of the sensor sensitivity. Considering high sensitivity and low fabrication complexity, the dual core PCF temperature sensor may be a better option to monitor or check the temperature of manufacturing industry, medical environment, transformer oil, battery of electric vehicles and so on.","PeriodicalId":6721,"journal":{"name":"2020 IEEE Region 10 Symposium (TENSYMP)","volume":"1 1","pages":"1764-1769"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89197600","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/TENSYMP50017.2020.9230952
M. Al Mamun, M. Istiak, Khandakar Abdulla Al Mamun, Sharifa Akter Rukaia
Wireless power transfer (WPT) using magnetic resonance, the technology which could set human free from the annoying wires. At present different researches have been taken place and are going on to increase the efficiency of the wireless power transfer. Researcher are trying to increase the distance between the transmitter and the receiver with a greater transfer power efficiency. This project implemented WPT innovation to charge the battery of electric vehicle. In this investigation, the distance level between transmitter and receiver circuit has been optimized, and at the same time the different power level at different distances between transmitter and receiver examined. Henceforth, different conditions of current and voltage levels from input to output circuit for delivering power to the battery which is completely took place without using any wire between transmitter circuits of charging station to receiver circuit of the electric vehicle for finding its output efficiency. Finally, its costs and other factors have been discussed to implement this wireless technology among electric vehicles. Several output levels in accordance with its input levels during the charging time has been recorded this empirical findings helps to solve several problems appears in wired technology such as electric shocks, cost, hassles due to wire, charging procedure.
{"title":"Design and Implementation of A Wireless Charging System for Electric Vehicles","authors":"M. Al Mamun, M. Istiak, Khandakar Abdulla Al Mamun, Sharifa Akter Rukaia","doi":"10.1109/TENSYMP50017.2020.9230952","DOIUrl":"https://doi.org/10.1109/TENSYMP50017.2020.9230952","url":null,"abstract":"Wireless power transfer (WPT) using magnetic resonance, the technology which could set human free from the annoying wires. At present different researches have been taken place and are going on to increase the efficiency of the wireless power transfer. Researcher are trying to increase the distance between the transmitter and the receiver with a greater transfer power efficiency. This project implemented WPT innovation to charge the battery of electric vehicle. In this investigation, the distance level between transmitter and receiver circuit has been optimized, and at the same time the different power level at different distances between transmitter and receiver examined. Henceforth, different conditions of current and voltage levels from input to output circuit for delivering power to the battery which is completely took place without using any wire between transmitter circuits of charging station to receiver circuit of the electric vehicle for finding its output efficiency. Finally, its costs and other factors have been discussed to implement this wireless technology among electric vehicles. Several output levels in accordance with its input levels during the charging time has been recorded this empirical findings helps to solve several problems appears in wired technology such as electric shocks, cost, hassles due to wire, charging procedure.","PeriodicalId":6721,"journal":{"name":"2020 IEEE Region 10 Symposium (TENSYMP)","volume":"35 1","pages":"504-507"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89231988","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/TENSYMP50017.2020.9230700
M. A. Islam, J. R. Mou, Md. Faruk Hossain, M. S. Hossain
In this paper, the structural, optical and photoluminescence (PL) features of the spray deposited Ba:SnO2 thin films with varying thickness have been investigated. X-ray diffraction analysis showed that Ba:SnO2 thin films have a polycrystalline tetragonal rutile structure and nanometric dimensions. The preferred directional growth of nanocrystalline was along <211> direction. The crystallite size of the samples was calculated using the Debye-Scherrer equation. With an increase in film thickness, the crystallite size decreased from 11.72 to 5.27 nm, while crystal imperfections, viz. lattice strain, degree of lattice distortion and dislocation density increased. The influence of film thickness on the transmittance and reflectance was investigated by using UV-Vis spectral data. The average visible transmittance of Ba:SnO2 nanocrystalline declined from 85 to 79% at 550 nm with an increase in film thickness. Ba:SnO2 nanocrystalline exhibited a slight red-shift in the absorption edge and the optical band gap was found to decrease in the range 3.94 to 3.64 eV. Near band-edge UV emissions band at ~ 402 nm were observed in the PL spectra which are attributed to the defect levels originating from the oxygen vacancies.
{"title":"Near-band edge UV emission and band gap engineering of highly transparent Ba:SnO2 nanocrystalline thin films","authors":"M. A. Islam, J. R. Mou, Md. Faruk Hossain, M. S. Hossain","doi":"10.1109/TENSYMP50017.2020.9230700","DOIUrl":"https://doi.org/10.1109/TENSYMP50017.2020.9230700","url":null,"abstract":"In this paper, the structural, optical and photoluminescence (PL) features of the spray deposited Ba:SnO2 thin films with varying thickness have been investigated. X-ray diffraction analysis showed that Ba:SnO2 thin films have a polycrystalline tetragonal rutile structure and nanometric dimensions. The preferred directional growth of nanocrystalline was along <211> direction. The crystallite size of the samples was calculated using the Debye-Scherrer equation. With an increase in film thickness, the crystallite size decreased from 11.72 to 5.27 nm, while crystal imperfections, viz. lattice strain, degree of lattice distortion and dislocation density increased. The influence of film thickness on the transmittance and reflectance was investigated by using UV-Vis spectral data. The average visible transmittance of Ba:SnO2 nanocrystalline declined from 85 to 79% at 550 nm with an increase in film thickness. Ba:SnO2 nanocrystalline exhibited a slight red-shift in the absorption edge and the optical band gap was found to decrease in the range 3.94 to 3.64 eV. Near band-edge UV emissions band at ~ 402 nm were observed in the PL spectra which are attributed to the defect levels originating from the oxygen vacancies.","PeriodicalId":6721,"journal":{"name":"2020 IEEE Region 10 Symposium (TENSYMP)","volume":"7 1","pages":"1552-1555"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89458871","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/TENSYMP50017.2020.9230638
S. Akter, Mohammad Shahriar Rahman, N. Mansoor
This paper introduces an efficient reactive routing protocol considering the mobility and the reliability of a node in Cognitive Radio Sensor Networks (CRSNs). The proposed protocol accommodates the dynamic behavior of the spectrum availability and selects a stable transmission path from a source node to the destination. Outlined as a weighted graph problem, the proposed protocol measures the weight for an edge the measuring the mobility patterns of the nodes and channel availability. Furthermore, the mobility pattern of a node is defined in the proposed routing protocol from the viewpoint of distance, speed, direction, and node's reliability. Besides, the spectrum awareness in the proposed protocol is measured over the number of shared common channels and the channel quality. It is anticipated that the proposed protocol shows efficient routing performance by selecting stable and secured paths from source to destination. Simulation is carried out to assess the performance of the protocol where it is witnessed that the proposed routing protocol outperforms existing ones.
{"title":"An Efficient Routing Protocol for Secured Communication in Cognitive Radio Sensor Networks","authors":"S. Akter, Mohammad Shahriar Rahman, N. Mansoor","doi":"10.1109/TENSYMP50017.2020.9230638","DOIUrl":"https://doi.org/10.1109/TENSYMP50017.2020.9230638","url":null,"abstract":"This paper introduces an efficient reactive routing protocol considering the mobility and the reliability of a node in Cognitive Radio Sensor Networks (CRSNs). The proposed protocol accommodates the dynamic behavior of the spectrum availability and selects a stable transmission path from a source node to the destination. Outlined as a weighted graph problem, the proposed protocol measures the weight for an edge the measuring the mobility patterns of the nodes and channel availability. Furthermore, the mobility pattern of a node is defined in the proposed routing protocol from the viewpoint of distance, speed, direction, and node's reliability. Besides, the spectrum awareness in the proposed protocol is measured over the number of shared common channels and the channel quality. It is anticipated that the proposed protocol shows efficient routing performance by selecting stable and secured paths from source to destination. Simulation is carried out to assess the performance of the protocol where it is witnessed that the proposed routing protocol outperforms existing ones.","PeriodicalId":6721,"journal":{"name":"2020 IEEE Region 10 Symposium (TENSYMP)","volume":"70 1","pages":"1713-1716"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89514713","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/TENSYMP50017.2020.9230785
Salma Tabashum, M. M. Hossain, Md. Ariful Islam, Mun Yea Mahafi Taz Zahara, Fahmida Naznin Fami
This work is dedicated to Bangla Crime Type Classification. As very few works had been done for Bangla crime classifier. To carry out this research, first we have developed a Bangla crime dataset which contains around 24,295 news articles and made most of them publicly available at github. Then we have built our crime classifier model and trained the classifier with our own dataset. We have analyzed word vectors like bag of words, TF-IDF in state-of-art machine learning algorithms as well as most promising semantic and syntactic word embeddings like Word2Vec, GloVe, fast-Text in both shallow and deep CNN and RNN to select best word embeddings for our classifier module. Finally we have summarized the experimental result in tabular form. We can see that significant improved accuracy can be achieved using deep learning algorithms over state-of-art machine learning algorithms in classifying Bangla crime data. The final experimental result shows that using shallow CNN with fastText,proposed model is able to achieve 93.70% accuracy.
{"title":"Performance Analysis of Most Prominent Machine Learning and Deep Learning Algorithms In Classifying Bangla Crime News Articles","authors":"Salma Tabashum, M. M. Hossain, Md. Ariful Islam, Mun Yea Mahafi Taz Zahara, Fahmida Naznin Fami","doi":"10.1109/TENSYMP50017.2020.9230785","DOIUrl":"https://doi.org/10.1109/TENSYMP50017.2020.9230785","url":null,"abstract":"This work is dedicated to Bangla Crime Type Classification. As very few works had been done for Bangla crime classifier. To carry out this research, first we have developed a Bangla crime dataset which contains around 24,295 news articles and made most of them publicly available at github. Then we have built our crime classifier model and trained the classifier with our own dataset. We have analyzed word vectors like bag of words, TF-IDF in state-of-art machine learning algorithms as well as most promising semantic and syntactic word embeddings like Word2Vec, GloVe, fast-Text in both shallow and deep CNN and RNN to select best word embeddings for our classifier module. Finally we have summarized the experimental result in tabular form. We can see that significant improved accuracy can be achieved using deep learning algorithms over state-of-art machine learning algorithms in classifying Bangla crime data. The final experimental result shows that using shallow CNN with fastText,proposed model is able to achieve 93.70% accuracy.","PeriodicalId":6721,"journal":{"name":"2020 IEEE Region 10 Symposium (TENSYMP)","volume":"48 1","pages":"1273-1277"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86617001","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/TENSYMP50017.2020.9230600
Md. Mahmudul Hasan Sabbir, Abu Sayeed, Md. Ahsan-Uz-Zaman Jamee
Diabetic Retinopathy, one of the dominant causes of vision loss to millions worldwide, can be prevented by early detection through regular retinal screening. People in less developed areas do not have adequate access to proper screening system because of their financial limitations. A cost-effective computer-aided screening system is presented in this paper using retinal fundus image. Ensemble learning helps to enhance the accuracy of the system by combining predictions of several learning models. In addition, these models are trained on texture features derived from gray level co-occurrence matrix (GLCM) as they are more effective to determine patterns from any images. Publicly available MESSIDOR fundus image dataset is used for experimental validation and the final results show that voting-based ensemble learning method with texture features achieves 97.2% sensitivity, 78.6% specificity and 92.0% accuracy which is higher than any individual learning model.
{"title":"Diabetic Retinopathy Detection using Texture Features and Ensemble Learning","authors":"Md. Mahmudul Hasan Sabbir, Abu Sayeed, Md. Ahsan-Uz-Zaman Jamee","doi":"10.1109/TENSYMP50017.2020.9230600","DOIUrl":"https://doi.org/10.1109/TENSYMP50017.2020.9230600","url":null,"abstract":"Diabetic Retinopathy, one of the dominant causes of vision loss to millions worldwide, can be prevented by early detection through regular retinal screening. People in less developed areas do not have adequate access to proper screening system because of their financial limitations. A cost-effective computer-aided screening system is presented in this paper using retinal fundus image. Ensemble learning helps to enhance the accuracy of the system by combining predictions of several learning models. In addition, these models are trained on texture features derived from gray level co-occurrence matrix (GLCM) as they are more effective to determine patterns from any images. Publicly available MESSIDOR fundus image dataset is used for experimental validation and the final results show that voting-based ensemble learning method with texture features achieves 97.2% sensitivity, 78.6% specificity and 92.0% accuracy which is higher than any individual learning model.","PeriodicalId":6721,"journal":{"name":"2020 IEEE Region 10 Symposium (TENSYMP)","volume":"68 1","pages":"178-181"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85795969","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：